How online marketing works and how to optimize budgets automatically. Same presentation given in two different data science events in Helsinki this week.
Guest lecture given in introductory data science course in Aalto University on 2017-11-30. About how data science is used in online marketing by Facebook and Smartly.io
1. Facebook uses an auction system called Vickrey-Clarke-Groves to determine ad placement and pricing, where advertisers bid for placement but pay close to the second highest bid to maximize their own value and optimize the consumer experience.
2. Advertisers should bid their "true value" which is the average lifetime profit per conversion minus the Facebook marketing cost, as this bidding strategy allows Facebook to optimize placement for both advertisers and consumers.
3. When estimating true value and setting bids, advertisers should consider factors like viral effects, repeat purchases, attribution windows beyond just one-day clicks, and view-through conversions to avoid underestimating their true value.
When marketing teams spend money on a paid acquisitions program it is crucial to understand the effect of that ad spend. In this talk, we will outline incrementality as a way to measure the causal impact that ad spend has on acquiring new customers and its advantages over more traditional metrics. We will walk through several ad measurement products available today and give examples of how to apply them to your business.
Acing the A/B Test: How to Optimize to Send Extraordinary EmailsMarketo
Emails have been around since the time of dinosaurs (maybe just the late 70s, but you get the point) - yet, nobody has ever sent the perfect email. While the perfect email might not even exist, how can you begin to create better emails with each and every send? Watch Tim Ozmina, Sr. Marketing Specialist, for his webinar, Acing the A/B Test: How to Optimize to Send Extraordinary Emails, where he dives into the world of email A/B testing.
You'll learn:
-What you should A/B test in your emails
-How to set up a proper A/B test
-How to accurately report on your results and make decisions
How online marketing works and how to optimize budgets automatically. Same presentation given in two different data science events in Helsinki this week.
Guest lecture given in introductory data science course in Aalto University on 2017-11-30. About how data science is used in online marketing by Facebook and Smartly.io
1. Facebook uses an auction system called Vickrey-Clarke-Groves to determine ad placement and pricing, where advertisers bid for placement but pay close to the second highest bid to maximize their own value and optimize the consumer experience.
2. Advertisers should bid their "true value" which is the average lifetime profit per conversion minus the Facebook marketing cost, as this bidding strategy allows Facebook to optimize placement for both advertisers and consumers.
3. When estimating true value and setting bids, advertisers should consider factors like viral effects, repeat purchases, attribution windows beyond just one-day clicks, and view-through conversions to avoid underestimating their true value.
When marketing teams spend money on a paid acquisitions program it is crucial to understand the effect of that ad spend. In this talk, we will outline incrementality as a way to measure the causal impact that ad spend has on acquiring new customers and its advantages over more traditional metrics. We will walk through several ad measurement products available today and give examples of how to apply them to your business.
Acing the A/B Test: How to Optimize to Send Extraordinary EmailsMarketo
Emails have been around since the time of dinosaurs (maybe just the late 70s, but you get the point) - yet, nobody has ever sent the perfect email. While the perfect email might not even exist, how can you begin to create better emails with each and every send? Watch Tim Ozmina, Sr. Marketing Specialist, for his webinar, Acing the A/B Test: How to Optimize to Send Extraordinary Emails, where he dives into the world of email A/B testing.
You'll learn:
-What you should A/B test in your emails
-How to set up a proper A/B test
-How to accurately report on your results and make decisions
Brent Chudoba of SurveyMonkey gave a presentation on lifetime value (LTV) to marketing students. He discussed how companies can calculate the lifetime revenue (LTR) of a customer by multiplying average monthly subscription cost by average number of months subscribed. For example, using $10/month for Netflix and an average subscription length of 18 months yields an LTR of $180. Companies aim to spend less than this on customer acquisition via marketing.
In-app offers monetization from basics to advanced technics based on Crazy Pa...GameCamp
From simple user segmentation to a complex interconnected system, we will follow Crazy Panda`s progress in developing personalized monetization. The key message of the session is not only about when or who to show those offers to, but also about how to make them truly personal. Presentation delivered by Ivan Kozyev, Head of Analytics of Crazy Panda at 8th edition of GameCamp (www.GameCamp.io).
Testing Makes Perfect: A Case Study in Onsite OptimizationsVe Interactive, US
Ve Interactive, US and Avangate deep-dive into a client case study and discuss the onsite optimizations that drove a 34% mo/mo revenue increase for client, Kingsoft.
This document discusses lessons learned about email marketing testing. It recommends testing design elements, timing, frequency, copy and images to understand audiences, increase click-through rates and returns. It describes A/B testing of different email elements and multivariate testing using real-time optimization. A case study showed prize testing increased unique views by 51% and personalization increased conversions by over 800%. The key lesson is that untested campaigns are guessing, so start testing to improve performance.
A comprehensive summary about the lifetime value predictions that Martin, Head of Marketing at AppAgent, has learned by building analytics platforms for clients and consulting with the best of the best in the mobile industry.
Maximize Customer Value (Optimove @ Market King Meetup - August 26, 2014)Amit Bivas
Retention Marketing: Maximize Your Customers' Value
Slides presented by Optimove's Head of Matrketing - Amit Bivas, at the Market King Meetup, August 26, 2014, Tel-Aviv.
Michael Twomey, Silverpop - 'Building ROI'iStrategy
iStrategy Conference Melbourne 2011: Michael Twomey, Silverpop, 'From Social Media to Email | Building your company's ROI' - Interactive Workshop (Day One)
This document discusses strategies for building a $100,000/month SaaS business, including:
1. Pursuing large total addressable markets (TAMs) over $1 billion or focusing on niche markets under $200 million.
2. Using funding strategically when it can power growth, but recognizing execution is also key.
3. Leveraging low costs in India to improve features, marketing, customer success.
4. Optimizing the full customer funnel around acquisition, activation, retention, revenue, and referral/growth.
5. Achieving product-market fit through solving customer problems and reaching appropriate scale before focusing on growth.
Three case studies about using GDN (Google Display Network)
1) How to evaluate impact of banners ads?
2) New format HTML5 and its oustanding results?
3) How to target B2B with banner ads?
The document discusses several applications of big data analysis including predicting movie ratings for Netflix, reducing customer churn for wireless providers, and targeting web advertisements. For each application, the analysis found limitations in prediction accuracy, with errors still existing even with complex algorithms. The key lessons are that while big data can provide insights, all applications will have some level of error and managers should not create undue hype but rather focus on using data-driven insights to inform marketing strategies and budgets.
Goals and Goal Funnels | Presented by Justin Spencer of Analytics ProsCaleb Whitmore
This document discusses best practices for setting goals and analyzing funnels in Google Analytics. It covers types of goals like URL, event, and engagement based goals. It also discusses using goals to extract insights and setting strategic goal architectures. Funnel analysis and best practices for setting up goal flows are also covered. The conclusion emphasizes that goals are a powerful analytics tool when set up strategically.
This document discusses the importance of long-term brand building versus short-term sales activation strategies in marketing. It argues that short-term strategies may produce quick sales uplifts but do not lead to sustainable long-term growth. Effective long-term strategies focus on reducing price sensitivity through branding, which produces slower but longer-lasting sales growth and increased profit margins over time. However, the trend in marketing is toward more short-termism as organizations prioritize quick gains over long-term effectiveness. The document advocates for an approach that views marketing investments as long-term investments, focuses on desired outcomes rather than just outputs, and considers how purchasing behaviors impact the value received.
amfAR, The Foundation for AIDS Research, took home a Direct Marketing Fundraisers Association (DMFA) Package of the Year award in the Email Fundraising category. SankyNet Vice President Paul Habig gave this great presentation on 9/28 to showcase the campaign's success.
The Private Life of Mail at Advertising Week Europe - A critique by Les Binet...Royal Mail MarketReach
If you attended Advertising Week Europe 2015, you will have heard Les Binet, Head of Effectiveness, adam&eveDDB, and Peter Field, renowned author on media effectiveness, critique the results of The Private Life of Mail; 18 months of unprecedented research into mail in the home, heart and head.
Insights into how the digital age impacts on mail integrating effectiveness, measurement and planning. Everything you thought about mail, may not be true. Learn what they think this means for the way mail is planned and measured in the digital age.
Presentation OMExpo - mobile and acquisition emailPowerspace
- Pwspace is a full-service email advertising solutions company launched in 2008 with 30 employees across offices in Barcelona and Paris, operating in 7 markets and serving 45 active clients and 180 publishers.
- Mobile usage continues to grow significantly in key European markets, with over 55% of connections in countries like the UK, Spain, Germany, France and Italy now coming from smartphones.
- Email remains the top mobile activity, with 50% of mobile opens and 52% of leads coming from emails on smartphones and tablets. However, only 20% of advertisers currently have emails that are fully optimized for mobile.
The document outlines strategies for testing email marketing campaigns during an economic downturn. It recommends developing a testing plan with short, mid, and long-term objectives. Short-term tests could explore offers, products, subject lines, frequency, creative content and personalization. Mid-term tests should build out the customer lifecycle and automate programs. Long-term tests may expand email's integration with other channels and introduce loyalty programs. Testing is presented as a low-cost way to adapt communications, boost revenue and improve ROI during uncertain times.
Leveraging Content to Generate Leads: 3 simple tactics one company used to ac...MarketingExperiments
Content marketing is a popular marketing strategy. Many marketers spend a significant amount of time developing content to generate interest from their customers.
But how do you use this content to generate a direct response? When you do, how do you ensure that you get the most response from the content you have created?
As one company recently discovered, it isn’t as simple as just putting a piece of content on your website. In fact, you could have valuable content to offer (as they did), but if you are unable to communicate this content in an effective way, then you are likely wasting much of the effort put into creating that content.
In a recent experiment with one of the largest industrial product distributers in the world, we were able to generate a 96% increase in the total number of leads by simply making a few tweaks to the way a piece of content was presented to the customer. We didn’t change, alter or rewrite the content. We simply ensured the expression of its value was felt at its fullest force.
In this 35-minute Web clinic, the MECLABS research team explained three simple tactics that you can employ on your own content marketing campaigns.
Presentation from CIMCIG Digital Marketing Workshop in March 2011. This presentation focuses on measuring what matters for your website.....outcomes or leads and understanding conversion rates.
Going Beyond Loyalty Programs to Retain Your Customers – Optimove - GrowComme...GrowCommerce
This document discusses strategies for retaining customers beyond basic loyalty programs. It emphasizes using data science and precision marketing techniques to foster emotional loyalty. Key points include segmenting customers using migration patterns, building predictive models tailored to specific needs, designing experiments to test individual messages and communication streams, and catering marketing to each customer's unique journey over time. The goal is to achieve the "ultimate" in customer marketing through speed, emotional intelligence, and a dynamic, data-driven approach.
Adventures in Business Analytics – Optimization and the Organization Garry, s...Tin Ho
Adventures in Business
Analytics – Optimization
and the Organization
Steve Garry
Marketing Optimization and the Organization
November 2014
Generating Better Business
Results Through Analytics
Brent Chudoba of SurveyMonkey gave a presentation on lifetime value (LTV) to marketing students. He discussed how companies can calculate the lifetime revenue (LTR) of a customer by multiplying average monthly subscription cost by average number of months subscribed. For example, using $10/month for Netflix and an average subscription length of 18 months yields an LTR of $180. Companies aim to spend less than this on customer acquisition via marketing.
In-app offers monetization from basics to advanced technics based on Crazy Pa...GameCamp
From simple user segmentation to a complex interconnected system, we will follow Crazy Panda`s progress in developing personalized monetization. The key message of the session is not only about when or who to show those offers to, but also about how to make them truly personal. Presentation delivered by Ivan Kozyev, Head of Analytics of Crazy Panda at 8th edition of GameCamp (www.GameCamp.io).
Testing Makes Perfect: A Case Study in Onsite OptimizationsVe Interactive, US
Ve Interactive, US and Avangate deep-dive into a client case study and discuss the onsite optimizations that drove a 34% mo/mo revenue increase for client, Kingsoft.
This document discusses lessons learned about email marketing testing. It recommends testing design elements, timing, frequency, copy and images to understand audiences, increase click-through rates and returns. It describes A/B testing of different email elements and multivariate testing using real-time optimization. A case study showed prize testing increased unique views by 51% and personalization increased conversions by over 800%. The key lesson is that untested campaigns are guessing, so start testing to improve performance.
A comprehensive summary about the lifetime value predictions that Martin, Head of Marketing at AppAgent, has learned by building analytics platforms for clients and consulting with the best of the best in the mobile industry.
Maximize Customer Value (Optimove @ Market King Meetup - August 26, 2014)Amit Bivas
Retention Marketing: Maximize Your Customers' Value
Slides presented by Optimove's Head of Matrketing - Amit Bivas, at the Market King Meetup, August 26, 2014, Tel-Aviv.
Michael Twomey, Silverpop - 'Building ROI'iStrategy
iStrategy Conference Melbourne 2011: Michael Twomey, Silverpop, 'From Social Media to Email | Building your company's ROI' - Interactive Workshop (Day One)
This document discusses strategies for building a $100,000/month SaaS business, including:
1. Pursuing large total addressable markets (TAMs) over $1 billion or focusing on niche markets under $200 million.
2. Using funding strategically when it can power growth, but recognizing execution is also key.
3. Leveraging low costs in India to improve features, marketing, customer success.
4. Optimizing the full customer funnel around acquisition, activation, retention, revenue, and referral/growth.
5. Achieving product-market fit through solving customer problems and reaching appropriate scale before focusing on growth.
Three case studies about using GDN (Google Display Network)
1) How to evaluate impact of banners ads?
2) New format HTML5 and its oustanding results?
3) How to target B2B with banner ads?
The document discusses several applications of big data analysis including predicting movie ratings for Netflix, reducing customer churn for wireless providers, and targeting web advertisements. For each application, the analysis found limitations in prediction accuracy, with errors still existing even with complex algorithms. The key lessons are that while big data can provide insights, all applications will have some level of error and managers should not create undue hype but rather focus on using data-driven insights to inform marketing strategies and budgets.
Goals and Goal Funnels | Presented by Justin Spencer of Analytics ProsCaleb Whitmore
This document discusses best practices for setting goals and analyzing funnels in Google Analytics. It covers types of goals like URL, event, and engagement based goals. It also discusses using goals to extract insights and setting strategic goal architectures. Funnel analysis and best practices for setting up goal flows are also covered. The conclusion emphasizes that goals are a powerful analytics tool when set up strategically.
This document discusses the importance of long-term brand building versus short-term sales activation strategies in marketing. It argues that short-term strategies may produce quick sales uplifts but do not lead to sustainable long-term growth. Effective long-term strategies focus on reducing price sensitivity through branding, which produces slower but longer-lasting sales growth and increased profit margins over time. However, the trend in marketing is toward more short-termism as organizations prioritize quick gains over long-term effectiveness. The document advocates for an approach that views marketing investments as long-term investments, focuses on desired outcomes rather than just outputs, and considers how purchasing behaviors impact the value received.
amfAR, The Foundation for AIDS Research, took home a Direct Marketing Fundraisers Association (DMFA) Package of the Year award in the Email Fundraising category. SankyNet Vice President Paul Habig gave this great presentation on 9/28 to showcase the campaign's success.
The Private Life of Mail at Advertising Week Europe - A critique by Les Binet...Royal Mail MarketReach
If you attended Advertising Week Europe 2015, you will have heard Les Binet, Head of Effectiveness, adam&eveDDB, and Peter Field, renowned author on media effectiveness, critique the results of The Private Life of Mail; 18 months of unprecedented research into mail in the home, heart and head.
Insights into how the digital age impacts on mail integrating effectiveness, measurement and planning. Everything you thought about mail, may not be true. Learn what they think this means for the way mail is planned and measured in the digital age.
Presentation OMExpo - mobile and acquisition emailPowerspace
- Pwspace is a full-service email advertising solutions company launched in 2008 with 30 employees across offices in Barcelona and Paris, operating in 7 markets and serving 45 active clients and 180 publishers.
- Mobile usage continues to grow significantly in key European markets, with over 55% of connections in countries like the UK, Spain, Germany, France and Italy now coming from smartphones.
- Email remains the top mobile activity, with 50% of mobile opens and 52% of leads coming from emails on smartphones and tablets. However, only 20% of advertisers currently have emails that are fully optimized for mobile.
The document outlines strategies for testing email marketing campaigns during an economic downturn. It recommends developing a testing plan with short, mid, and long-term objectives. Short-term tests could explore offers, products, subject lines, frequency, creative content and personalization. Mid-term tests should build out the customer lifecycle and automate programs. Long-term tests may expand email's integration with other channels and introduce loyalty programs. Testing is presented as a low-cost way to adapt communications, boost revenue and improve ROI during uncertain times.
Leveraging Content to Generate Leads: 3 simple tactics one company used to ac...MarketingExperiments
Content marketing is a popular marketing strategy. Many marketers spend a significant amount of time developing content to generate interest from their customers.
But how do you use this content to generate a direct response? When you do, how do you ensure that you get the most response from the content you have created?
As one company recently discovered, it isn’t as simple as just putting a piece of content on your website. In fact, you could have valuable content to offer (as they did), but if you are unable to communicate this content in an effective way, then you are likely wasting much of the effort put into creating that content.
In a recent experiment with one of the largest industrial product distributers in the world, we were able to generate a 96% increase in the total number of leads by simply making a few tweaks to the way a piece of content was presented to the customer. We didn’t change, alter or rewrite the content. We simply ensured the expression of its value was felt at its fullest force.
In this 35-minute Web clinic, the MECLABS research team explained three simple tactics that you can employ on your own content marketing campaigns.
Presentation from CIMCIG Digital Marketing Workshop in March 2011. This presentation focuses on measuring what matters for your website.....outcomes or leads and understanding conversion rates.
Going Beyond Loyalty Programs to Retain Your Customers – Optimove - GrowComme...GrowCommerce
This document discusses strategies for retaining customers beyond basic loyalty programs. It emphasizes using data science and precision marketing techniques to foster emotional loyalty. Key points include segmenting customers using migration patterns, building predictive models tailored to specific needs, designing experiments to test individual messages and communication streams, and catering marketing to each customer's unique journey over time. The goal is to achieve the "ultimate" in customer marketing through speed, emotional intelligence, and a dynamic, data-driven approach.
Adventures in Business Analytics – Optimization and the Organization Garry, s...Tin Ho
Adventures in Business
Analytics – Optimization
and the Organization
Steve Garry
Marketing Optimization and the Organization
November 2014
Generating Better Business
Results Through Analytics
This document summarizes how an analytics/AI/BPO company called TechM can help address issues in the e-commerce sector. It discusses the buyer and seller cycles in e-commerce and how TechM can use analytics and automation to improve processes. For buyers, TechM can help verify customers and route transactions appropriately based on risk scores. For sellers, TechM can help with onboarding, pricing optimization, and culling bad sellers. TechM can also help e-commerce companies better manage their supply chains and returns through predictive analytics and an early warning system.
Simon Harrison gave a presentation on the analytics lifecycle. He discussed descriptive, diagnostic, predictive, and prescriptive analytics. Descriptive analytics explains what happened, diagnostic identifies why outcomes occurred, predictive determines what will happen, and prescriptive optimizes outcomes subject to constraints. Harrison explained how each stage helps businesses and provided examples of analytics products that can be used. He also discussed how analytics impacts accounting skills and how organizations can approach their analytics journey.
Lean advertising 101 - Paid Acquisition for StartupsClément Delangue
This document provides an overview of lean advertising for startups, specifically paid acquisition. It discusses why entrepreneurs may be hesitant to try paid acquisition and argues that it can help startups test and reach product-market fit faster. The document then provides a 3-step process for startups to scale paid acquisition campaigns within a week, including setting up analytics, experimenting with different channels spending $5-15 per day, and iterating and scaling what performs best. It concludes by highlighting some common mistakes to avoid, such as optimizing for the wrong goals or trusting aesthetics over metrics.
Google Analytics is a powerful tool, but it has some limitations, particularly with very large data. Thankfully, Google has provided an excellent API to allow access to your data- and with tools like Analytics Canvas companies are pushing what they can do with their Google Analytics data to the limit.
Mark Roberge is a Senior Lecturer with Harvard Business School, former CRO of Hubspot and author of bestseller "The Sales Acceleration Formula". Join him as he takes you through his step by step guide to revenue growth.
This document discusses using machine learning to analyze IoT data and calculate the return on investment (ROI) for various use cases. It provides an example ROI case study of using machine learning in manufacturing to increase overall equipment effectiveness (OEE) and lower production costs. The summary outlines how machine learning can be used to predict and prevent quality issues, analyze diagnostic data, and plan for higher quality at lower variable costs. It estimates that plant managers could increase OEE by up to 6% and achieve a target ROI of 175% with payback within one year.
Let’s make web analytics more customer centric!Optimics s.r.o.
This document discusses making web analytics more customer-centric. It argues that current tools focus too much on simple metrics like bounce rate and conversion rate that do not capture customer relationships and behavior over time. The document advocates focusing on metrics like annual retention rates, new customer counts, customer lifetime value, and how customers migrate between channels over time. It provides an example analysis of how paid search is better for acquiring but not retaining customers long-term. The analysis joined order data, campaign data, and traffic sources to understand customer channel preferences and profitability. The goal is to optimize marketing spend based on expected long-term customer revenue rather than short-term conversions.
Lessons learned from 3k SaaS Companies - Patrick Campbell, CEO, Price Intelli...SaaStock
Patrick Campbell, CEO, Price Intelligently presented "Lessons learned from 3k SaaS Companies" on the Playbook Stage at SaaStock 2016.
Watch the video here: https://www.saastock.com/blog/view/lessons-learned-from-3k-saas-companies
SaaStock is Europe’s premier Conference for B2B SaaS at RDS, Dublin Ireland in September 2017 www.saastock.com
Better Living Through Analytics - Strategies for Data DecisionsProduct School
Data is king! Get ready to understand how a successful analytics team can empower managers from product, marketing, and other areas to make effective, data-driven decisions.
Louis Cialdella, a data scientist at ZipRecruiter, shared some case studies and successful strategies that he has used at ZipRecruiter as well as previous experiences. The purpose of this data talk was to enlighten people on how to make sure that analysts can successfully partner with other departments and get them the information they need to do great things.
Intro to Data Analytics with Oscar's Director of ProductProduct School
The Director of Product at Oscar, Vasudev Vadlamudi, went over key types of quantitative analysis that B2C product managers use on the job including: funnels, cohorts, and a/b testing. For each one he looked into when and why they are used, and used examples.
[Webinar] Innovate Faster by Adopting The Modern Growth StackOptimizely
Adopting a growth mindset requires that product teams become experts in their customers’ behavior. But often legacy technologies simply count clicks and users, and fail to provide insights that product teams need to accelerate growth.
Attend this webinar and learn how the modern growth stack can help you:
-Overcome obstacles that often prevent enterprises from moving quickly
-Use experimentation and product analytics to reduce uncertainty and increase data-driven decisions
-Design a Modern Growth Stack built from best of breed solutions to accelerate product innovation
This webinar is part of our Change the Game series.
Content marketing analytics: how to make your data work harder for your businessDaniel Smulevich
Presentation from Search London - July 2014.
The goal of our work is to understand the triggers behind new and existing customers content consumption and buying behaviour so we can plot the best course of action to drive demonstrable improvements in revenue.
Here is a 3-step process to do that.
Class 03: Introduction to Google AnalyticsJon Chang
This document provides an introduction to Google Analytics. It includes learning objectives around key web analytics terms, metrics, best practices, and features in Google Analytics like goals, segments, and campaign tracking. The document covers setting up important conversions and segmentation to understand user journeys and how to improve conversion rates. It emphasizes establishing normal ranges for key metrics and investigating exceptions to determine opportunities.
Social Media ROI: Formulas & Conversion Optimization TipsAngie Schottmuller
Are YOU measuring social media ROI? Now you can! Loaded with 14 formulas for quantifying social media benefits and over 17 case study examples of social proof facilitating conversion, this presentation is a MUST-READ for any social marketer!
Still not convinced of the impact of social media on sales? Test it with the Unbounce landing page testing platform: http://try.unbounce.com/seschi (Use promo code ses503 to get 3 months at 50% off.)
Presented on Nov 13, 2012 at SES Chicago, #SESCHI.
This document summarizes a presentation on Trade Promotion Intelligence. It discusses the challenges of current data repository systems and benefits of an ideal state. It then reviews the TradeSmart architecture, which automates data integration and allows users to access harmonized data. Several case studies are presented that show how the TradeSmart system can generate insightful reports and analyses to evaluate promotion performance and ROI.
How to Use Data for Product Decisions by YouTube Product ManagerProduct School
With millions of new data points every moment, how are Product Managers expected to make sense of it all? This talk outlined the steps required to distill and synthesize data to drive actionable product decisions. The most effective Product Managers are those who know their data: they can justify product priority and roadmap changes, calibrate resource asks and manage their own time more effectively. This lecture equipped the audience with the tools necessary to draw insight from unstructured data using Google’s cloud analytics suite.
This document discusses automating work by connecting Google Analytics to spreadsheets. It introduces a suite of six products for automating analytics and growth: Funnel Analysis for automated multi-funnel reporting; Live Results for real-time insights delivered by email; Test Impact Analysis for calculating test effects; Experiment Analysis for automated multi-segment A/B test analysis; Sampling Fixer for adjusting test samples; and Quick Insights for impressions of customer behavior. Instructions are provided on setting up the Google Analytics Sheets API and using it to retrieve and analyze custom data requests in spreadsheets for flexible reporting and multiple use cases.
Spike it’s an online platform that makes AI accessible for all businesses, governments and other enterprises.
Every business needs artificial intelligence.
But not every business has the capability to build, maintain, and generate value from AI in-house.
Spike makes AI accessible for all businesses, governments and other enterprises.
By harnessing the power of your data to assist – not displace – humans at work, we improve business efficiency and productivity.
And all of that will be provided with:
1- mine data warehouse
2- Interactive Analytics
3- Prediction and forecasting report.
With:
* No AI or machine learning experience required.
* No infrastructure needed.
* Supporting the MENA region and we are less expensive for other solutions.
Similar to Running Stan in Production: Bayesian Revenue Estimation (20)
The Ipsos - AI - Monitor 2024 Report.pdfSocial Samosa
According to Ipsos AI Monitor's 2024 report, 65% Indians said that products and services using AI have profoundly changed their daily life in the past 3-5 years.
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"sameer shah
Embark on a captivating financial journey with 'Financial Odyssey,' our hackathon project. Delve deep into the past performance of two companies as we employ an array of financial statement analysis techniques. From ratio analysis to trend analysis, uncover insights crucial for informed decision-making in the dynamic world of finance."
STATATHON: Unleashing the Power of Statistics in a 48-Hour Knowledge Extravag...sameer shah
"Join us for STATATHON, a dynamic 2-day event dedicated to exploring statistical knowledge and its real-world applications. From theory to practice, participants engage in intensive learning sessions, workshops, and challenges, fostering a deeper understanding of statistical methodologies and their significance in various fields."
End-to-end pipeline agility - Berlin Buzzwords 2024Lars Albertsson
We describe how we achieve high change agility in data engineering by eliminating the fear of breaking downstream data pipelines through end-to-end pipeline testing, and by using schema metaprogramming to safely eliminate boilerplate involved in changes that affect whole pipelines.
A quick poll on agility in changing pipelines from end to end indicated a huge span in capabilities. For the question "How long time does it take for all downstream pipelines to be adapted to an upstream change," the median response was 6 months, but some respondents could do it in less than a day. When quantitative data engineering differences between the best and worst are measured, the span is often 100x-1000x, sometimes even more.
A long time ago, we suffered at Spotify from fear of changing pipelines due to not knowing what the impact might be downstream. We made plans for a technical solution to test pipelines end-to-end to mitigate that fear, but the effort failed for cultural reasons. We eventually solved this challenge, but in a different context. In this presentation we will describe how we test full pipelines effectively by manipulating workflow orchestration, which enables us to make changes in pipelines without fear of breaking downstream.
Making schema changes that affect many jobs also involves a lot of toil and boilerplate. Using schema-on-read mitigates some of it, but has drawbacks since it makes it more difficult to detect errors early. We will describe how we have rejected this tradeoff by applying schema metaprogramming, eliminating boilerplate but keeping the protection of static typing, thereby further improving agility to quickly modify data pipelines without fear.
Running Stan in Production: Bayesian Revenue Estimation
1. Running Stan in
Production: Bayesian
Revenue Estimation
2018-08-29 StanCon
Markus Ojala
PhD, Chief Data Scientist
Smartly.io
@MarkusOjala
2. Implemented in early 2017:
https://www.smartly.io/blog/tutorial-how-we-productized-bay
esian-revenue-estimation-with-stan
Referred by Andrew Gelman in early 2018:
http://andrewgelman.com/2018/01/21/smartly-io-productize
d-bayesian-revenue-estimation-stan/
“It’s BDA come to life!“
“This sort of thing is exactly what we were hoping to see.”
Blog posts
3. 1 billion spend
Managed yearly
We Make Online Advertising
Easy, Effective, And
Enjoyable
Facebook &
Instagram Partner
7. The number of pulls for a given lever should match
its actual probability of being the optimal lever
Sample from the posterior for the mean of each lever
Bayesian Bandits / Thompson sampling
9. Separate revenue model into two parts
ROAS = revenue / cost
= revenue / conversions * conversions / cost
= revenue / conversions * 1 / CPA
ROAS = return on ad spend
CPA = cost per action
Existing model
● Lot of data
● Varies fast
● Big differences
New model
● Little data
● Varies slowly
● Small real differences
● Lot of random variation
10. ● Use hierarchical model: Account -> Campaign -> Ad set
● Revenue follows usually a long-tailed distribution, use log-normal
● We observe only hourly aggregates: approximate by using
Fenton-Wilkinson log-normal approximation
Modeling the revenue per purchase
16. ● Easy way to write Bayesian models and do inference
● But hard to get stable in production: limit the scope of the model
● Reparametrize, use reasonable informative priors and custom initialization
○ Solves most of the convergence issues
● Sampling does not scale: we use ADVI / variational inference approximation
○ Validate that you get same results with sampling and ADVI
○ Random initialization issues were solved by fixing ADVI parameters
● About 1000 daily runs and $1M scheduled with Celery, PyStan, monitoring
Learnings in productizing Stan
21. We observe only aggregates
Goal: estimate log-normal parameters for ad sets
Challenge: observation i is aggregate of multiple events ni
Solution: Estimate by another log-normal